Two years at the Computational Cognitive Science discussion group: A student’s perspective

From left to right: Alex Doumas (Senior Lecturer, Psychology) Sydelle de Souza (Doctoral Researcher, CDT in NLP, School of Informatics), Naiti Bhatt (PhD Student, Psychology), Jean Skelton (PhD Student, Psychology), Erica Adezati (PhD Student, Psychology).
Photo credit: Ivan Vegner. (From left to right: Alex Doumas (Senior Lecturer, Psychology) Sydelle de Souza (Doctoral Researcher, CDT in NLP, School of Informatics), Naiti Bhatt (PhD Student, Psychology), Jean Skelton (PhD Student, Psychology), Erica Adezati (PhD Student, Psychology)).

In this post, Sydelle shares the incredible journey of their discussion group, which has evolved into a close-knit community, eroding hierarchies and fostering a collegiate learning environment. Sydelle de Souza, a doctoral researcher at the School of Informatics, elaborates on their Computational Cognitive Science discussion group, funded by the IAD Student Partnership Agreement↗️. This post belongs to the Hot topic series: Student Partnership Agreement 2023↗️

The University of Edinburgh boasts a rich history of excellence in cognitive science (CogSci). In fact, the term “cognitive science” itself was coined at this very institution in 1973. Over the years, Edinburgh has celebrated numerous pioneering achievements in the field, including hosting the first international meeting of the Cognitive Science Society outside the US, in 2001. These milestones were catalysed by the university’s unique culture, which encourages a free flow of ideas across traditionally separate disciplines such as psychology, neuroscience, artificial intelligence (AI), philosophy, and linguistics. 

Although the university has become more diverse and inclusive, the CogSci community found itself with limited opportunities for interaction between staff and students, especially across departments. To counter this trend, in the academic year 2021/2022, an initiative was launched at the School of Philosophy, Psychology and Language Sciences (PPLS)↗️. The aim was to create a meta-lab of sorts wherein students and lecturers could engage with each other beyond the confines of the classroom. This resulted in the Computational Cognitive Science Reading Discussion Group (CCSDG). 

With commanding views of Edinburgh Castle, room 7.01 of the Dugald Stewart Building played host to fortnightly meetings of our discussion group. A collaborative organizational effort between students and staff, our reading group embarked on a journey of intellectual exploration and critical engagement with topics in CogSci. We welcomed external speakers, dissected seminal papers, and provided an engaging atmosphere for exchanging ideas, replete with pizza. Owing to funding from the Student Partnership Agreement (SPA), our fortnightly meetings became weekly in 2022/2023 and our membership saw an increase in numbers as well as in diversity. Postgraduate students from Informatics, Psychology, and Linguistics joined us every week as we engaged in lively debate on topics both longstanding and current.  

From causal inference and theory of mind to reinforcement learning and large language models, members proposed empirical papers and review articles they found intriguing and thought-provoking. The proposals came from students, postdoctoral fellows, and professors alike. At the meeting, the proposer led the discussion, offering everyone a chance to delve deeply into the methodologies, results, and implications of each paper. Observing senior academics dissect a paper and formulate their arguments not only helped us hone our skills in critical evaluation and scientific inquiry, but also gave us an appreciation of the intricacies of academic discourse and how to conduct ourselves. Most importantly, we were exposed to cutting-edge research that we would not have seen in the classroom or through the narrow lens of our niche research areas. We honed our critical thinking skills by discussing trending topics at the nexus of CogSci and AI; these discussions proved to be especially relevant given the rapid societal changes propagated by advances in AI. In so doing, we learned to truly appreciate the interdisciplinary nature of cognitive science and its potential for impact in the real world. 

A significant highlight of our reading group was the opportunity to interact with a remarkable array of external speakers, including Andrew Lampinen, Senior Research Scientist at Google’s Deep Mind (↗️) and Charley Wu (↗️), PI of the Human and Machine Cognition Lab at the University of Tübingen in Germany. The SPA funding allowed us to host them in Edinburgh and engage with them in person. These talks generated quite a bit of excitement at PPLS and Informatics, and we saw several new faces attend (a few of whom have joined our ranks this year). The Q&A sessions that followed the talks provided us with valuable insights, sparked stimulating debates, and broadened our perspectives. As a bonus, the focus on inviting early career researchers gave us a sneak peek into what our future careers as cognitive scientists could look like.  

As the year wore on, our discussion group evolved into a close-knit community. It was noticeable that students were more at ease interacting with professors, voicing their opinions, and asking questions. Conversely, our senior members also gained a deeper understanding of the diverse perspectives of their students. The discussion group became a space where hierarchies were blurred, and traditional academic boundaries transcended. We had created an environment for Edinburgh CogSci researchers that was both intellectually stimulating and socially rewarding. Our plan had come to fruition. 

Now, in November 2023, our discussion group continues to thrive. While many of our student members have graduated, we have welcomed several new faces. We meet every fortnight on Thursday from 1200 to 1300 in room 7.01 at the Dugald Stewart Building. If you’re a CogSci, or a CogSci-adjacent researcher, or just someone who is interested in CogSci and are looking for community, come join us. We’re a friendly bunch, plus there’s pizza! 

Photograph of the authorSydelle de Souza

Sydelle de Souza is a UKRI-funded doctoral researcher at the CDT in Natural Language Processing, School of Informatics. She aspires to be a computational cognitive scientist and is broadly interested in the analogical mechanisms at play in language acquisition. Her current work investigates how domain-general cognitive mechanisms, data-driven learning, and cultural transmission can account for the emergence of linguistic structures with idiosyncratic meanings in language. Syd holds an MSc in Psychological Research from the University of Edinburgh, an MSc in Applied Linguistics and Second Language Acquisition from the University of Oxford, and a BA in Language Sciences from the University of Lisbon.

Leave a Reply

Your email address will not be published. Required fields are marked *